I'm a big fan of voice dictation apps. In fact, I'm using one right now to write this very post (you'll have to wait till the end to see which one I'm using )

The two main products I've used in this space are @Aqua Voice and @Wispr Flow. From talking to others, these are the two that I typically hear people mention using. In general, I hear a lot more people talk about using Wispr Flow.

AI dev tools are moving stupid fast. Every few weeks, there s a new must-use. Some stick. Most don t.

Some vibe coders are developing full products with @ChatGPT by OpenAI+ @Replit. Others swear by @Cursor + @Claude by Anthropic . A few are mixing @Lovable , @v0 by Vercel , and @bolt.new . New and shipping way faster than expected.

I ve been refining my own vibe stack lately.
Building with @Google Antigravity at the core. It keeps the flow clean when things get messy.

Share your current Vibe Stack:

I ve been noticing something interesting while working with AI tools.

Give two engineers the same task and the same AI, and the results can be completely different.

Not because one is better overall, but because of how they prompt.

Some are very structured and clear, others are vague, and the output reflects that.

I keep seeing advice like use this model for the easy stuff and that one for complex problems. But it makes me wonder what really counts as a complex problem for an LLM?

For us, complex usually means lots of steps, deep reasoning, or tricky knowledge. But for AI, the definition might be different. Some things that feel easy for us can be surprisingly hard for models, while things that seem tough for us (like scanning huge datasets quickly) might be trivial for them.

I was inspired by @chrismessina to build my own Raycast extensions after he released his own! I have been using Raycast a lot more thanks to a gentle nudge from @jakecrump (and @Raycast's latest AI releases), so I knew I wanted to build my own Raycast extension at some point.

There was a lot of news about Cyberpunk for Mac released and I wanted to search if there was a deal for it on PC. I instinctively went to search Raycast .but it wasn t there. So my extension idea was born!

My goal was simple: make it easier to hunt for video game deals right from the Raycast command bar. What followed was a rapid journey of prototyping, debugging, and learning. I m excited to share what I learned and hope it will encourage more makers to ship their ideas.
You can view my IsThereAnyDeal (ITAD) Extension on GitHub here! I will launch it when it's approved on the Raycast store :) I'll publish a more in-depth, nerdy guide later. But for now...here's my experience building a Raycast Extension!

a week ago i asked one question here: how do you actually keep up with AI without drowning. i expected tool recommendations. i got something better.

almost nobody talked about which tools. they talked about how to learn at all. the patterns that kept coming up:

saving isnt learning. a bookmark is a parking lot, not a to-do list. it feels like progress and nothing sticks until you actually need it.

stop trying to learn every tool. the people who seemed sane picked a few tied to real work and let the rest go. let the project tell you which tools matter.

I've been vibe coding for a few months as a non-coder and I'm still annoyed with the fact that I can't understand what's happening under the hood.

I've got a decent understanding of code but I can't actually write it and I don't know the best practices for stuff like architecture and security which apparently aren't baked into most vibe coding tools. So my question is...

if i use claude or codex, i constantly am screen capping to show bad padding / alignment / whatever. is there a defacto way to let claude or codex tool call to see the browser or render a page for themselves?

For me, it s not generating UI or features anymore.

It s maintaining consistency as the product becomes larger:

• architecture

• state management

• debugging

• scaling

• avoiding messy code generation loops

• keeping UX coherent across pages

We've been getting the same request over and over from our users: "My AI gives great answers, but it can't actually do anything."

It got us thinking most AI integrations today are still essentially fancy search boxes. The AI talks, the human acts. But the real unlock is when the AI can close the loop itself query the database, send the email, update the record without a human in the middle.

The hard part isn't the action itself. It's the non-determinism. How do you build a system where the AI decides when to act, which action to take, and what parameters to pass based purely on context without it going off the rails?

A few things we've learned building this: